Artificial neural network models to predict nodal status in clinically node-negative breast cancer

Dihge, Looket; Ohlsson, Mattias; Edén, Patrik; Bendahl, Pär Ola; Rydén, Lisa

doi:10.1186/s12885-019-5827-6

Cited by 32 publications

(112 citation statements)

References 49 publications

Supporting

Mentioning

109

Contrasting

Order By: Relevance

“…ANNs are analytical techniques that have been successful in solving classification problems in different domains [26][27][28][29][30]. Based on the functioning of biological neural networks, ANNs are dense networks of interconnected artificial neurons that get activated based on inputs.…”

Section: Predictive Modelsmentioning

confidence: 99%

Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches

Tarekegn¹,

Ricceri²,

Costa³

et al. 2020

JMIR Med Inform

View full text Add to dashboard Cite

Background Frailty is one of the most critical age-related conditions in older adults. It is often recognized as a syndrome of physiological decline in late life, characterized by a marked vulnerability to adverse health outcomes. A clear operational definition of frailty, however, has not been agreed so far. There is a wide range of studies on the detection of frailty and their association with mortality. Several of these studies have focused on the possible risk factors associated with frailty in the elderly population while predicting who will be at increased risk of frailty is still overlooked in clinical settings. Objective The objective of our study was to develop predictive models for frailty conditions in older people using different machine learning methods based on a database of clinical characteristics and socioeconomic factors. Methods An administrative health database containing 1,095,612 elderly people aged 65 or older with 58 input variables and 6 output variables was used. We first identify and define six problems/outputs as surrogates of frailty. We then resolve the imbalanced nature of the data through resampling process and a comparative study between the different machine learning (ML) algorithms – Artificial neural network (ANN), Genetic programming (GP), Support vector machines (SVM), Random Forest (RF), Logistic regression (LR) and Decision tree (DT) – was carried out. The performance of each model was evaluated using a separate unseen dataset. Results Predicting mortality outcome has shown higher performance with ANN (TPR 0.81, TNR 0.76, accuracy 0.78, F1-score 0.79) and SVM (TPR 0.77, TNR 0.80, accuracy 0.79, F1-score 0.78) than predicting the other outcomes. On average, over the six problems, the DT classifier has shown the lowest accuracy, while other models (GP, LR, RF, ANN, and SVM) performed better. All models have shown lower accuracy in predicting an event of an emergency admission with red code than predicting fracture and disability. In predicting urgent hospitalization, only SVM achieved better performance (TPR 0.75, TNR 0.77, accuracy 0.73, F1-score 0.76) with the 10-fold cross validation compared with other models in all evaluation metrics. Conclusions We developed machine learning models for predicting frailty conditions (mortality, urgent hospitalization, disability, fracture, and emergency admission). The results show that the prediction performance of machine learning models significantly varies from problem to problem in terms of different evaluation metrics. Through further improvement, the model that performs better can be used as a base for developing decision-support tools to improve early identification and prediction of frail older adults.

show abstract

Section: Predictive Modelsmentioning

confidence: 99%

Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches

Tarekegn¹,

Ricceri²,

Costa³

et al. 2020

JMIR Med Inform

View full text Add to dashboard Cite

show abstract

“…The highest true validation value (AUC: 0.75) suggests that the SUS nomogram does not perfectly predict a disease-free axilla, which highlights the complexity of lymphatic spread. Although more complex prediction models may have certain advantages for estimating nodal involvement 44 , 45 , the SUS nomogram is a readily available and user-friendly predictive tool in clinical settings. Studies based on retrospective registry data may be considered unreliable, given the risks of incomplete or improperly recorded data.…”

Section: Discussionmentioning

confidence: 99%

Validation of the Skåne University Hospital nomogram for the preoperative prediction of a disease-free axilla in patients with breast cancer

et al. 2021

View full text Add to dashboard Cite

Background Axillary staging via sentinel lymph node biopsy (SLNB) is performed for clinically node-negative (N0) breast cancer patients. The Skåne University Hospital (SUS) nomogram was developed to assess the possibility of omitting SLNB for patients with a low risk of nodal metastasis. Area under the receiver operating characteristic curve (AUC) was 0.74. The aim was to validate the SUS nomogram using only routinely collected data from the Swedish National Quality Registry for Breast Cancer at two breast cancer centres during different time periods. Method This retrospective study included patients with primary breast cancer who were treated at centres in Lund and Malmö during 2008–2013. Clinicopathological predictors in the SUS nomogram were age, mode of detection, tumour size, multifocality, lymphovascular invasion and surrogate molecular subtype. Multiple imputation was used for missing data. Validation performance was assessed using AUC and calibration. Results The study included 2939 patients (1318 patients treated in Lund and 1621 treated in Malmö). Node-positive disease was detected in 1008 patients. The overall validation AUC was 0.74 (Lund cohort AUC: 0.75, Malmö cohort AUC: 0.73), and the calibration was satisfactory. Accepting a false-negative rate of 5 per cent for predicting N0, a possible SLNB reduction rate of 15 per cent was obtained in the overall cohort. Conclusion The SUS nomogram provided acceptable power for predicting a disease-free axilla in the validation cohort. This tool may assist surgeons in identifying and counselling patients with a low risk of nodal metastasis on the omission of SLNB staging.

show abstract

“…The NILS prediction model for nodal prediction was originally constructed, including ten top ranked risk variables for nodal status (tumor size, vascular invasion, multifocality, estrogen receptor status, histological type, progesterone receptor status, mode of detection, age, tumor localization in the breast, and Ki-67 positivity) (10). The validation study will validate the original model, including the above-mentioned variables.…”

Section: Methodsmentioning

confidence: 99%

The NILS study protocol - a retrospective validation study of a preoperative decision-making tool for non-invasive lymph node staging in women with primary breast cancer [ISRCTN14341750]

Skarping

Dihge

Bendahl

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Background Routine preoperative axillary ultrasonography has proven insufficient for detecting low-burden nodal metastatic deposits. For the majority of newly diagnosed breast cancer patients presenting with clinical T1-T2 N0 disease, the standard axillary staging by sentinel lymph node biopsy is not therapeutic. The pilot non-invasive lymph node staging (NILS) artificial neural network (ANN) model to predict nodal status was published in 2019. The aim of the current study is to assess the performance measures of the model for the prediction of healthy lymph nodes in clinically N0 breast cancer patients at two breast cancer centers in Sweden. Methods This bicenter, observational, retrospective study has been designed to validate the NILS prediction model for nodal status using preoperatively collected clinicopathological and radiological data. A web-based implementation of the nodal status classifier has been developed and will be used in this study, resulting in an estimated probability of healthy lymph nodes for each study participant. Our primary endpoint is to report on the performance of the NILS prediction model to distinguish between healthy and metastatic lymph nodes (discrimination, N0 vs. N+) and compare the observed and predicted event rates of benign axillary nodal status (calibration). Discussion Internationally, there are numerous artificial intelligence projects involving non-invasive identification of N0 breast cancer. Here, we present a robust validation study based on external cohorts of our ANN model. Although validation is necessary to show generalizability, it is often overlooked. If the accuracy and discrimination reach a satisfactory level, our prediction tool can be implemented to assist medical professionals and breast cancer patients in shared decision-making on omitting sentinel node biopsy in patients predicted to be node-negative. In future, this may potentially save healthcare resources and reduce costs and adverse side effects. In addition, our study might prompt future studies of nodal metastases of malignancies in other organs, and thus might have implications beyond breast cancer. Trial registration This study has been prospectively registered in the ISRCTN registry, identification number: 14341750

show abstract

Artificial neural network models to predict nodal status in clinically node-negative breast cancer

Cited by 32 publications

References 49 publications

Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches

Predictive Modeling for Frailty Conditions in Elderly People: Machine Learning Approaches

Validation of the Skåne University Hospital nomogram for the preoperative prediction of a disease-free axilla in patients with breast cancer

The NILS study protocol - a retrospective validation study of a preoperative decision-making tool for non-invasive lymph node staging in women with primary breast cancer [ISRCTN14341750]

Contact Info

Product

Resources

About